Dollars and Jens
Monday, September 12, 2011
Not Statistically Significant Does Not Mean Zero
I know I've seen this mistake in an accounting paper or two, but I haven't seen it a lot.
In theory, a comparison of two experimental effects requires a statistical test on their difference. In practice, this comparison is often based on an incorrect procedure involving two separate tests in which researchers conclude that effects differ when one effect is significant (P < 0.05) but the other is not (P > 0.05). We reviewed 513 behavioral, systems and cognitive neuroscience articles in five top-ranking journals (Science, Nature, Nature Neuroscience, Neuron and The Journal of Neuroscience) and found that 78 used the correct procedure and 79 used the incorrect procedure. An additional analysis suggests that incorrect analyses of interactions are even more common in cellular and molecular neuroscience. We discuss scenarios in which the erroneous procedure is particularly beguiling.I suppose real scientists don't have to understand statistics as well as we do because they can do controlled experiments, so the statistics they need is a lot simpler.